March 1, 2013
Scientists Electronically Connect Rat Brains To Create ‘Organic Computer’
Alan McStravick for redOrbit.com - Your Universe Online
In an article written earlier this month for redOrbit, this writer highlighted work being done in the Nicolelis laboratory headed by professor of neurobiology Miguel Nicolelis at Duke University School of Medicine. His previous research reported on the ability to enable a rat to “touch” infrared light through the use of a cortical stimulation technique.For the second time this month, the Nicolelis team has released an equally fascinating report that builds on their previous research. He and his team have managed to electronically link the brains of pairs of rats for the first time. The methods used show amazing possibilities that Nicolelis himself states could be modified for human use in the not-too-distant future. And because it seems that the possibilities for this kind of technology could soon be truly be limitless, we might do ourselves and our posterity a service by exploring the ethical questions that arise from the success of Nicolelis´ work.
The study, published today in the journal Scientific Reports, offers results that suggest the near-future potential for linking together several brains to form what the researchers refer to as an “organic computer” that could potentially allow for the sharing of motor and sensory information among groups of animals.
"Our previous studies with brain-machine interfaces had convinced us that the rat brain was much more plastic than we had previously thought," said Nicolelis, lead author of the study. "In those experiments, the rat brain was able to adapt easily to accept input from devices outside the body and even learn how to process invisible infrared light generated by an artificial sensor. So, the question we asked was, 'if the brain could assimilate signals from artificial sensors, could it also assimilate information input from sensors from a different body?'"
CREATING AN “ORGANIC COMPUTER”
Before the team could begin their research, they first had to train pairs of rats to press the correct lever when an indicator light above the lever switched on. A reward system was implemented whereby the rats would receive a sip of water if they performed the action correctly. Once this training was complete, they were then able to connect two animals´ brains using arrays of microelectrodes with a thickness of about 1/100th of a human hair. These tiny electrodes were inserted into the area of the cortex responsible for processing motor information.
At this point, each pair was divided into two separate designations. The first animal, the “encoder,” was presented with the visual cue that guided it to the correct lever to receive its water reward. The actions of the encoder were then translated into a pattern of electrical stimulation by sampling its brain activity that coded its behavioral decisions. The second rat, known as the “decoder,” then received the electrical stimulation pattern directly to its brain via the microelectrodes.
In the decoder rat´s enclosure were the same types of levers as were found in the encoder´s enclosure. The only difference was that the visual cue was not present in the decoder rat´s cage. By eliminating the visual cue, the research team ensured that if the decoder rat wanted to receive its reward, it would have to rely on the cue being transmitted from the encoder via the brain-to-brain interface.
A series of trials were conducted to determine the ability of the decoder rat to receive and interpret the brain input being sent by the encoder rat and to press the correct lever in order to receive a reward. After 45 days of one-hour practice sessions with the animals, the researchers noted the success rate of their rat pairs was at about 70 percent — about 8 percentage points below what researchers had theorized as the maximum success rate achievable based on success rates of sending signals directly to the decoder rat´s brain.
Proficiency among the decoder rats to receive and interpret the signal from their encoder counterpart, as noted above, came after 45 days. According to the research team, it was on day 45 that there seemed to be a sudden switch in understanding among the decoder rats. As Nicolelis described it, “there is a moment in time when “¦ it clicks. Suddenly, the [decoder] animal realizes ℠Oops! The solution is in my head. It´s coming to me´ and he gets it right.”
COLLABORATION BETWEEN LINKED BRAINS
For the encoder rat, his new environment was different from the initial reward-training scenario in only one aspect: His reward was denied each time his decoder rat partner selected the wrong lever in his own environment. The research team believed that this might force the decoder animal to send clearer, more intense thought signals to the encoder. It would, according to Licolelis, create a “behavioral collaboration” between the two animals.
"We saw that when the decoder rat committed an error, the encoder basically changed both its brain function and behavior to make it easier for its partner to get it right," Nicolelis said. "The encoder improved the signal-to-noise ratio of its brain activity that represented the decision, so the signal became cleaner and easier to detect. And it made a quicker, cleaner decision to choose the correct lever to press. Invariably, when the encoder made those adaptations, the decoder got the right decision more often, so they both got a better reward."
The researchers also conducted a second set of experiments in which the rat pairs were trained to distinguish between narrow and wide openings using their whiskers. The reward system used by the team for this experiment trained the rats to seek their reward on the left side of the chamber if the opening was narrow and, conversely, on the right side of the chamber if the opening was wide.
Much like the first part of the experiment, the pairs were again separated into decoders and encoders. The decoders were trained to associate stimulation pulsations with a reward administered on the left side of their chamber. If no pulsation was felt, the decoder knew to seek reward on the right side of the chamber. The reported success rate for this experiment was about 65 percent which, the team points out, is significantly higher than what would be expected from pure chance.
“It is important to stress that the topology of BTBI [Brain-to-Brain Interface] does not need to be restricted to one encoder and one decoder subjects. Instead, we have already proposed that, in theory, channel accuracy can be increased if instead of a dyad whole grids of multiple reciprocally interconnected brains are employed. Such a computing structure could define the first example of an organic computer capable of solving heuristic problems that would be deemed non-computable by a general Turing-machine” Nicolelis explained.
To take their work on BTBI one step further, Nicolelis and colleagues at the Edmond and Lily Safra International Institute of Neuroscience of Natal (ELS-IINN) in Brazil separated an encoder and decoder duo by thousands of miles of land and sea. The brain signals of an encoder in Brazil were transmitted to the decoder in North Carolina using the Internet. The results showed that the two rats were still able to work together on the tactile discrimination task.
"So, even though the animals were on different continents, with the resulting noisy transmission and signal delays, they could still communicate," said Miguel Pais-Vieira, PhD, a postdoctoral fellow and co-author of the study. "This tells us that it could be possible to create a workable, network of animal brains distributed in many different locations."
Nicolelis added that their experiments “demonstrated the ability to establish a sophisticated, direct communication linkage between rat brains, and that the decoder brain is working as a pattern-recognition device. So basically, we are creating an organic computer that solves a puzzle.”
"But in this case, we are not inputting instructions, but rather only a signal that represents a decision made by the encoder, which is transmitted to the decoder's brain which has to figure out how to solve the puzzle. So, we are creating a single central nervous system made up of two rat brains."
THE ETHICS OF THE “BRAIN NET” AND NEUROPHYSIOLOGICAL SOCIAL INTERACTIONS
Nicolelis also explained that his system does not need to be limited to only a single pair of brains but rather could include a network of brains or, as he idiosyncratically calls it, a “brain net.” His team in North Carolina, in partnership with his colleagues at the ELS-IINN, are currently exploring new opportunities for experimentation that will enable them to link multiple animals cooperatively to solve more complex behavioral tasks.
"We cannot predict what kinds of emergent properties would appear when animals begin interacting as part of a brain-net. In theory, you could imagine that a combination of brains could provide solutions that individual brains cannot achieve by themselves," continued Nicolelis. Such a connection might even mean that one animal would incorporate another's sense of "self," he said.
"In fact, our studies of the sensory cortex of the decoder rats in these experiments showed that the decoder's brain began to represent in its tactile cortex not only its own whiskers, but the encoder rat's whiskers, too. We detected cortical neurons that responded to both sets of whiskers, which means that the rat created a second representation of a second body on top of its own." Nicolelis believes his study of these adaptations will ultimately lead to a new field which he calls the “neurophysiology of social interaction.”
As noted by Professor Christopher James, an expert in neural engineering at the University of Warwick, there is simply no way to recreate this research in humans at this point in time using only the surface of the scalp.
“If you want to get information into the brain, then putting electrodes right at the brain sites is the way to do it. However, it´s clearly very invasive.”
It was the recognition of the invasive nature of the research that had James pondering the ethics of the research, itself. “It´s very, very interesting, isn´t it? Because in humans you´d obviously get informed consent in doing this.”
“It´s an exciting paper which basically shows that it is possible to take information out of the brain, and it is possible to take information and pump it into the brain. What this shows is that the technology is here. And the sort of things we should be talking about is: Why are we doing this, and what do we hope to get out of it?”
Lending his voice to this topic, Dr. Anders Sandberg, who studies the ethics of neurotechnologies at the Future of Humanity Institute at Oxford University acknowledged Nicolelis´ work was “very important” in helping us to understand how brains are able to encode information.
But the implications of the technology and its potential uses in the future can also be seen in the much broader context of human hegemony, said Sandberg. "The main reason we are running the planet is that we are amazingly good at communicating and coordinating. Without that, although we are very smart animals, we would not dominate the planet."
"I don't think there's any risk of supersmart rats from this," he added. "There's a big difference between sharing sensory information and being able to plan. I'm not worried about an imminent invasion of 'rat multiborgs'."
Very little is currently known about the biomechanics behind how thoughts are encoded and how they are transmitted into another person's brain. This is at least in part due to the fact that there is very often a chasm between what we think about doing and what we actually do. Much of what is in our minds is what Sandberg calls a "draft" of what we might do rather than concrete plans of action. "Often, we don't want to reveal those drafts, that would be embarrassing and confusing. And a lot of those drafts are changed before we act. Most of the time I think we'd be very thankful not to be in someone else's head."
In the formation of an “organic computer” the research team claims the complexity of the experiments will be enabled by the laboratory´s ability to take in and record the brain activity of nearly 2,000 brain cells at one time. They say that within five years time, the capability to simultaneously record upwards of 10-30,000 cortical neurons will have been achieved.
Nicolelis hopes that this work will help develop technology that will offer more precise control of motor neuroprostheses that help restore motor control to paralyzed individuals.
Just recently, Brazilian research funding agency FINEP awarded a research grant of $20 million to the Walk Again Project. With this grant, Walk Again will attempt to develop the first brain-controlled whole-body exoskeleton aimed at restoring mobility in severely paralyzed patients. Expect to see the first demonstration of this futuristic contraption sometime during the first game of the 2014 Soccer World Cup in Brazil.